The pyrrolo(1,4)benzodiazepine antitumor antibiotics, anthramycin, tomaymycin and sibiromycin are produced by various actinomycetes. While all three antibiotics have been shown to have a wide range of antitumor activity in animal studies, anthramycin and sibiromycin have also been shown to be active in human studies. The molecular basis for their cytotoxicity is due to their ability to form a covalent linkage with DNA, which results in inhibition of DNA and RNA synthesis. The overall objectives of the proposed research are: 1) to elucidate a general biosynthetic scheme for the biosynthesis of the pyrrolo(1,4)benzodiazepine antibiotics; 2) to elucidate the precise manner both in vitro and in vivo that these antitumor agents react with DNA; 3) to determine whether anthramycin plays a biological role for the producing organism; 4) to explore the possibility of administration of these antitumor agents as their DNA complexes, in order to circumvent various toxicities which limit the clinical application of these drugs. Our previous biosynthetic studies have established the biogenetic building blocks for these antibiotics. Studies are now planned using radioisotope and stable isotope techniques to establish the general pathway for conversion of tyrosine to the C2 and C3-proline units to these antibiotics and determine key intermediates in this pathway. The mode of action studies are aimed primarily at the isolation and structure determination of the antibiotic-nucleoside adduct(s) formed in vitro and comparison of these to the products formed in vivo by human cell lines. Preliminary data suggests that anthramycin may be implicated in the transition of the producing organism from growth phase to differentiation. Experiments are planned to determine whether anthramycin does bind to producer cell DNA and if by doing so, it may affect transcription and sporulation. Our in vitro studies on the antibiotic-DNA complexes have confirmed that under certain conditions free active antibiotics can be released from the complex. This observation has led us to examine the possible application of the drug-DNA complexes as pro-drugs of these antitumor agents. Preliminary indications lead us to suspect that the drug-DNA complexes are still active and the amount of tissue necrosis produced may be less than when the free drugs are administered.